PROJECT SUMMARY/ABSTRACT Broad Impact: Social anxiety disorder (SAD) is a prevalent mental health concern that impacts approximately 12% of the population. One mechanism thought to maintain SAD is avoidance of faces (i.e. avoidance of negative evaluative threat). However, research on attentional processes in SAD has been confined to paradigms presented on computer monitors. To investigate attentional processes in a more naturalistic way we developed an immersive, 360-video virtual reality environment using real actors, as part of a pilot study. Participants with a range of social anxiety symptoms (from none to severe) completed a 5-minute speech in this virtual reality environment while their eye movements were recorded. Results from the study showed that greater symptoms of social anxiety were associated with avoidance of looking at faces (i.e. fewer fixations on faces). While existing treatments for SAD are moderately effective, a large number of individuals do not experience meaningful reductions in their symptoms. The overarching goal of this project is to inform future treatment research for SAD. We will test a brief attention guidance intervention for SAD that specifically targets avoidance of faces as a potential mechanism maintaining the disorder. The proposed research will use the eye tracking hardware and naturalistic virtual reality environment from the pilot study. We will also collect eye tracking data prior to the intervention in order to investigate potential heterogeneity in the attentional processes of SAD. Aim 1 will test the hypotheses that (a) the attention guidance intervention, compared to the standard exposure intervention, will result in a greater reduction in symptoms of social anxiety, and (b) this effect will be mediated by the number of fixations on faces during a brief public speaking challenge post-intervention. These results will provide much needed evidence as to whether avoidance of faces operates as a causal maintaining factor in SAD. Aim 2 will test the hypothesis that there are distinct sub-groups of SAD with unique patterns of eye movements over the course of a pre-intervention speech collected as part of Aim 1. A Bayesian statistical approach will be implemented that evaluates patterns of eye movements and estimates group membership based on those patterns. This approach may support a more personalized approach to treating SAD. Taken together these findings may inform future experimental therapeutics for SAD utilizing targeted attentional processes. Training: This project will prepare the applicant to conduct early stage clinical trials utilizing brief targeted interventions aimed at putative causal mechanisms implicated in the maintenance of anxiety-related pathology. The applicant will also gain advanced technical, theoretical, and statistical knowledge critical for conducting eye movement (i.e. visual attention) research. This proposal will aid the applicant in disseminating his work to the scientific community through conference presentations and journal publications, as well as provide him with skills for his future academic career in clinical research.